Skip to content

Benchmark

Arena ELO - French

People compare model answers and choose which they prefer. Higher scores are better.

Models measured
262
Scores recorded
303
Sources
1
IQ weight
1.87

Score distribution

1,360.8
Mean
1,408.4
Median
127.5
Std dev
1,158.9
10th percentile
1,487.2
90th percentile
945.6
Min
945.61,562.9
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.87 for chat models, measured over 262 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
101Step 5 Preview· LMArena data1,444.5
102Grok 4.1· LMArena data1,443.7
103Mistral Large 4· LMArena data1,443.1
104GPT-5.4 miniHigh· LMArena data1,442.8
105Qwen3.5 122B-A10B· LMArena data1,442.6
106GPT-6 AstraMax· LMArena data1,441.4
107DeepSeek V4 ProHigh, 20260813· LMArena data1,440.5
108GPT-5.2 Chat20260210· LMArena data1,439.1
109DeepSeek V4 Flash· LMArena data1,439
110Inkling Small· LMArena data1,438.3
111GPT 5.2· LMArena data1,437.9
112MiniMax M3· LMArena data1,435.7
113DeepSeek-V4-Flash PreviewHigh· LMArena data1,435.5
114Gemma 4 31B IT· LMArena data1,435.5
115GPT 5.1· LMArena data1,433.9
116DeepSeek V3.2 Exp· LMArena data1,433.5
117Claude Opus 4.1Thinking 16K, 20250805· LMArena data1,432.3
118Gemini 2.5 Flash Preview09-2025· LMArena data1,431.9
119Gemini 2.5 Flash· LMArena data1,431.8
120Grok 4 FastReasoning· LMArena data1,431.6
121Grok 4.7xHigh· LMArena data1,430.8
122GLM 4.7· LMArena data1,430.4
123MiMo V2 FlashThinking· LMArena data1,429.7
124OpenAI o32025-04-16· LMArena data1,429.6
125DeepSeek V3.2· LMArena data1,429.5
126Claude Opus 4.120250805· LMArena data1,427.1
127nova-experimental-chat10-20· LMArena data1,426.9
128GPT-5.5 Instant· LMArena data1,426.4
129Kimi K2 Thinking· LMArena data1,425.5
130Claude Haiku 5.5High· LMArena data1,425.2
131MiniMax M2.7· LMArena data1,424.5
132Gemini 3.1 Flash Lite· LMArena data1,422.7
133DeepSeek V3.2Thinking· LMArena data1,422.1
134Step 3.5 Flash· LMArena data1,421.7
135MiMo V2 FlashNon-Thinking· LMArena data1,419.2
136Qwen3 30B A3B Instruct2507· LMArena data1,418.4
137GPT-4.5 Preview2025-02-27· LMArena data1,418.3
138GLM 4.5· LMArena data1,418
139Grok 40709· LMArena data1,417.5
140DeepSeek R10528· LMArena data1,415.6
141Grok 4.1 FastReasoning· LMArena data1,415.3
142Qwen3.5 Flash· LMArena data1,413.2
143LongCat-Flash-Chat Exp2602· LMArena data1,413.2
144Grok 4.3· LMArena data1,413.1
145Qwen3 Next 80B A3B Instruct· LMArena data1,413.1
146Qwen3.5 35B A3B· LMArena data1,412.7
147Claude Haiku 4.520251001· LMArena data1,411
148DeepSeek V3.2 ExpThinking· LMArena data1,410.5
149Qwen3.5 27B· LMArena data1,409.9
150Solar Pro 4· LMArena data1,409.3

Showing 101–150 of 303 results.

Data credits: LMArena, CC BY 4.0 (licence)